University of Groningen
Mode choice and mode commitment in commuters
Sivasubramaniyam, Rathee D. ; Charlton, Samuel G.; Sargisson, Rebecca J.
Published in:
Travel Behaviour and Society
DOI:
10.1016/j.tbs.2019.10.007
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Sivasubramaniyam, R. D., Charlton, S. G., & Sargisson, R. J. (2020). Mode choice and mode commitment
in commuters. Travel Behaviour and Society, 19, 20-32. https://doi.org/10.1016/j.tbs.2019.10.007
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Contents lists available atScienceDirect
Travel Behaviour and Society
journal homepage:www.elsevier.com/locate/tbs
Mode choice and mode commitment in commuters
Rathee D. Sivasubramaniyam
a,⁎, Samuel G. Charlton
a, Rebecca J. Sargisson
a,b aUniversity of Waikato, New ZealandbUniversity of Groningen, Netherlands
A R T I C L E I N F O Keywords: Mode choice Mode commitment Motivating factors Intention Travel diary Daily commute A B S T R A C T
In New Zealand, like many other developed countries, a majority of trips (67%) involve the use of private cars, producing negative effects on the environment and public health. Interventions aimed to reduce car use can be successful if we not only understand the reasons car users drive but also why other commuters use more sus-tainable alternatives. Although a range of possible motivating factors have been previously identified in the literature, the significance of the present study was to address the question of whether these motivating factors interact with each other to influence commuters’ intentions to choose a particular mode for their daily commute. A sample of commuters completed an online survey and a subset completed a 1-week travel diary later. Social norms were a significant predictor of drivers’ and car passengers’ intentions to use the car, whereas ease-of-use was a significant predictor of drivers’ intentions to drive and active commuters’ intentions to walk or cycle. All commuters had comparable ecological beliefs and mode-related status which were not related to their intentions to use their travel modes. Although all the commuters were committed to their mode choice for daily commute, drivers and pedestrians were more likely to use only their respective travel modes for daily commute, whereas passengers, bus users, and cyclists were more likely to use a combination of several modes. Future research might productively explore subtypes of car commuters and additional analysis techniques to identify ways of nudging car commuters to reduce their car use in favour of sustainable alternatives.
1. Introduction
A majority of commuter trips involve privately-owned cars and light vehicles and the number of vehicles and trips has continued to grow over time. The number of passenger cars in use globally has increased from 654 million in 2005 to 947 million in 2015 (OICA, 2015). In New Zealand, the number of passenger cars has increased since 2009 (3.1 million) with an accelerated growth from 2013 onwards (3.2 million) to the highest level ever recorded (3.8 million) in 2017. The amount of travel in passenger cars per capita has also continued to grow from 2012 with 8377 km of annual travel to 9265 km in 2017, also the highest ever recorded (Ministry of Transport, 2018). Unfortunately, the domination of car use over sustainable transport modes has negative consequences on the environment. Transportation is a major con-tributor of greenhouse gas emissions such as carbon dioxide (CO2),
nitrous oxide (N20), and methane (CH4) (Samet, 2007). Passenger cars
account for 45% of the total energy use and gas emissions in global transportation (Holmberg et al., 2012), and 57.7% in New Zealand transportation (Ministry of Transport, 2018). Therefore, there is a widely acknowledged need to encourage commuters to reduce their car use and increase their use of more sustainable modes such as taking the
bus, cycling, and walking.
Several researchers have examined how to encourage commuters to use more sustainable modes for their commute. Most of these attempts involved hard measures such as structural improvements by introdu-cing bus-only lanes and cycling paths (e.g., (Cervero et al., 2009; Jones and Sloman, 2003; Viegas and Lu, 2001) or service improvements by expanding the bus schedule and subsidising bus fares (e.g.,Haque et al., 2013; Jones and Sloman, 2003). However, Garling and Schuitema (2007) suggested that hard measures may be more effective if combined with soft measures, such as providing motivational support to change commuting goals, requesting plans for how to change, and providing customised information. Studies from the US and the Netherlands (e.g.,
Cairns et al., 2002; Shoup, 1997; Touwen, 1999) showed a 20–25% reduction in car use when the researchers combined soft with hard measures, achieving reductions of only 5–15% when only hard mea-sures were used (Cairns et al., 2008). To encourage commuters to change, policy makers and researchers must work together to develop an appropriate suite of measures that address commuters’ emotions, attitudes, social norms, personal norms, perceived responsibility, and awareness of the negative impacts of car use (Steg et al., 2001).
Researchers have examined a wide range of influences on
https://doi.org/10.1016/j.tbs.2019.10.007
Received 2 May 2019; Received in revised form 22 August 2019; Accepted 28 October 2019
⁎Corresponding author at: Transport Research Group, School of Psychology, University of Waikato, Private Bag 3105, Hamilton, New Zealand.
E-mail address:srd12@students.waikato.ac.nz(R.D. Sivasubramaniyam).
Available online 27 November 2019
2214-367X/ © 2019 Hong Kong Society for Transportation Studies. Published by Elsevier Ltd. All rights reserved.
commuting modes that can broadly describe types of commuters as: habitual commuters, economic commuters, pleasure-seeking ters, intentional commuters, and environmentally-conscious commu-ters. In practice, when commuters are faced with making the same trip every day, repetition in an invariant context may lead to a habitual and largely unconscious choice of travel mode. Repetition of a choice of mode leads to an increased likelihood of the same choice in the future (Bamberg et al., 2003a; Gardner, 2009; Ouellette and Wood, 1998). As with any habit, repetition may reduce conscious involvement in choosing a commuting mode. For example, involvement in a cogni-tively demanding task has been shown to make cyclists more likely to choose cycling as a transport mode, even after they had been asked not to choose cycling (as compared to cyclists who had not completed the task) (Aarts and Dijksterhuis, 2000). The demanding task diminished participants' mental resources available to select an alternative mode and they automatically selected their habitual travel mode. With ex-tended repetition, choice of travel mode can become stereotyped and context-bound such that contextual cues in familiar situations can au-tomatically trigger use of the habitual travel mode (Wood et al., 2005), allowing alternative choices to be considered only in unfamiliar or changed situations such as moving to a new residence (Bamberg, 2006), or a temporary expressway closure (Fujii and Garling, 2003).
Prior to the development of prepotent habits, commuters may choose a travel mode based on economic factors, explicitly considering the costs (time and money) of different modes. The decision-making literature suggests that costs are more salient than gains, and that people tend to be loss averse when making decisions because losses are more salient than gains (Kahneman and Tversky, 1979). In a change from usual car dominance, commuters will choose public transport when traffic is congested because it will take less time than driving (Cervero, 2002; Corpuz, 2007). During off-peak hours, however, com-muters were more likely to drive due to less frequent public transport schedules, and less traffic congestion (Corpuz, 2007). Congestion pri-cing policies have successfully leveraged commuters’ loss aversion to reduce car use (Eliasson et al., 2009; Washbrook et al., 2006). Similarly, the costs of car use (maintenance and fuel) are important considerations for some commuters that choose to use the bus (Beirao and Cabral, 2007).
Researchers have also shown that commuters’ choice of mode may be based on what makes them feel good or satisfied. Commuters may feel good because a mode fulfils their symbolic motives (e.g., sense of power, prestige, and status) and affective motives (e.g., autonomy, freedom, and control). For example, some commuters may choose to drive because they enjoy being in control of the vehicle and derive a sense of higher status when driving (Steg et al., 2001). Some cyclists find pleasure in improving their traffic skills and their fitness levels (Davies et al., 1997). Pedestrians often report a sense of relaxation and an absence of stress while walking during their commute (Anable and Gatersleben, 2005). Bus users report enjoying qualities specific to bus service such as accurate departure time, punctuality, reliability, and the opportunity to talk to other passengers (Ettema et al., 2012; Friman et al., 2001).
For some commuters, the decision to use a particular mode is rea-soned and intentional based on three social psychological factors: at-titudes, subjective norms, and perceived behavioural control (PBC).
Ajzen (1991) explored the relationship between intention and beha-viour and proposed in the Theory of Planned Behabeha-viour (TPB) that intentions to perform a behaviour should precede the behaviour itself.
Ajzen (1991) suggested that it is possible to influence behaviour and intention through interventions that significantly affect one or more of the three psychological factors. Several researchers have attempted to use TPB to encourage commuters to use the public transport using an intervention aimed at influencing commuters’ attitude, social norms, and PBC towards public transport (Bamberg et al., 2003a,b). After re-ceiving a prepaid bus ticket either alone (Bamberg et al., 2003a) or combined with an information pack (Bamberg et al., 2003b), students
were more likely to use public transport and had more positive atti-tudes, positive subjective norms, higher PBC, and stronger intentions to use public transport. Therefore, interventions targeted at influencing commuters’ attitudes, subjective norms, PBC, and intentions can be successful.
Finally, some commuters choose travel modes based on their pro-environmental attitudes. Environmentally conscious commuters may also be economic commuters to the extent they are loss averse with regard to the negative impacts that certain travel modes have on the environment (Flamm, 2009; Thogersen and Olander, 2006). Pro-en-vironmental commuters regularly perform pro-enPro-en-vironmental viours such as recycling and purchasing organic items. These beha-viours are consistent with more frequent use of public transport and reducing car use (Atasoy et al., 2012; Roberts et al., 2018; Thogersen and Olander, 2006; Whitmarsh and O'Neill, 2010). To the degree that pro-environmental attitudes are seen as social norms, commuters’ transport choices may also be shaped by these social and personal norms (Hunecke et al., 2001).
As can be seen in this brief review, there are a range of different types of considerations that may influence the commuting modes chosen by different types of commuters. Although there are different views on the degree to which these various considerations influence commuters to choose their travel modes, researchers have tended to examine the factors in isolation, such that each researcher focused on the role of just one or a very few factors influencing commuters’ mode choices. Conceptually this may be problematic both because different types of commuters may respond to different influences, and also be-cause commuters may be influenced by multiple factors simultaneously. The present study sought to address this issue by comparing the degree to which commuters’ intentions to use their typical travel mode is in-fluenced by a wide range of factors.
We wanted to directly compare the full range of factors that have been suggested to have an influence on commuters’ mode choices. Specifically, we identified measures to determine the relative influences of habits, economic decision-making, social norms, status, pleasure, ease-of-use, and ecological beliefs. Our motivation for doing this re-volved around the conceptual question of whether these seven potential motivating factors interact with each other to influence commuters’ intentions to choose a particular mode for their daily commute. As al-luded to earlier, the different sorts of findings previously reported in the literature opens the possibility that users of different commuting modes may have different factors motivating their intentions. We also ex-pected that for some commuters, only one factor may be influential whereas for other commuters, several factors may be influential, an aspect that has not previously been explored to any great degree.
In order to explore the relative influence of these different types of motivational factors we recruited a wide range of different types of commuters; those who regularly chose to drive their private car, use the bus, cycle, walk, or commute as a car passenger, across a wide range of demographic characteristics. We asked these commuters to complete an online survey and then selected a subset of them to complete a 1-week travel diary in which they provided details of their daily commute (e.g., trip duration, trip distance, and mode choice). We hoped to provide a fuller picture of how the range of underlying considerations interact to influence commuters’ intentions to choose a particular travel mode for their regular trips to work and school.
2. Method
The methodology for the present research involved recruiting a sample of commuters to complete an online survey followed by asking a subset of the respondents to complete a 1-week travel diary in which they provided details of their daily commute. The data collection for both the online survey and travel diary took place in New Zealand between March and May 2018 (Autumn in New Zealand).
2.1. Participants
We recruited respondents 16 years of age and older through notices placed on the intranets of various New Zealand organisations as well as on social media, news media, and through word-of-mouth. Five hun-dred and eighty-five respondents completed the online survey. The majority of them resided in a single region of New Zealand. We ex-cluded respondents who lived in other regions due to the low number of
responses and the availability of idiosyncratic travel modes unique to those areas within New Zealand (e.g., train and Uber). We also excluded respondents who did not use any of the five commuting modes of in-terest to us. The final regional sample of 313 respondents consisted of 239 women and 74 men, with a wide age range (Table 1). We cate-gorised respondents as drivers, car passengers, bus users, cyclists, or pedestrians based on their responses to one of the questions in the survey. The majority were drivers (n = 191), followed by pedestrians
Table 1
Demographic Characteristics of Commuters and Modes for Daily Commute (n = 313).
Percentage of respondents within each commuter group (%)
n Drivers (n = 191) Passengers (n = 22) Bus users (n = 24) Cyclists (n = 32) Pedestrians (n = 44) Gender Male 74 20.94 9.09 20.83 56.28 20.45 Female 239 79.06 90.91 79.17 43.75 79.55 Age 16–24 years 84 21.47 45.45 29.17 6.25 54.55 25–34 years 63 21.99 22.73 20.83 12.50 15.91 35–44 years 67 21.99 18.18 29.17 21.88 15.91 45–54 years 61 20.42 4.55 12.50 40.63 11.36 55–64 years 34 12.57 9.09 8.33 15.63 2.27 65–74 years 4 1.57 – – 3.13 – 75 years or over – – – – – – Level of education
No secondary school qualification 3 1.05 – – – 2.27
High school qualification or equivalent 84 22.51 40.91 33.33 9.38 47.73
Tertiary diplomas or certificate 58 20.42 13.64 29.17 9.38 13.64
Bachelor degree or Bachelor with Honours degree 112 42.41 9.09 20.83 40.63 25.00
Master degree or higher 56 13.61 36.36 16.67 40.63 11.36
Occupation*
Full time student 97 21.99 54.55 37.50 15.63 65.91
Part time student 27 10.47 4.55 12.50 6.25 2.27
Full time worker 183 65.97 40.91 50.00 68.75 31.82
Part time worker 56 19.37 18.18 12.5 18.75 13.64
Casual worker 16 4.71 13.64 4.17 – 6.82
Unemployed or looking for work 12 2.09 4.55 4.17 – 13.64
Looking after home or family 21 8.90 13.64 – 3.13 –
Retired 2 1.05 – – – –
Household type
Person living alone 24 7.33 – 4.17 6.25 15.91
Married/de facto couple 68 19.90 31.82 8.33 43.75 15.91
Other adults only (e.g., flatmates) 60 17.28 9.09 8.33 12.50 43.18
Family (including extended) with children 94 31.41 31.82 54.17 25.00 13.64
Family with adults only 49 15.71 27.27 25.00 12.50 6.82
Single adults living children 12 5.76 – – – 2.27
Other 6 2.62 – – – 2.27
Annual household income
Less than $50,000 82 18.85 27.27 33.33 9.38 65.91
$50,000 to $99,999 113 39.97 36.36 37.50 43.75 13.64
$100,000 to $149,999 65 23.56 22.73 8.33 21.88 13.64
$150,000 or more 48 16.23 9.09 20.83 25.00 4.55
Average daily travel time
Less than 10 min 24 4.19 18.18 – 12.50 18.18
10–19 min 79 27.23 22.73 8.33 25.00 27.27 20–29 min 69 20.42 22.73 12.50 28.13 29.55 30–39 min 39 13.61 4.55 20.83 12.50 6.82 40–49 min 37 13.09 9.09 25.00 6.25 4.55 50–59 min 24 6.28 13.64 16.67 6.25 6.82 60 min or over 41 15.18 9.09 16.67 9.38 6.82
Average daily travel distance
Less than 5 km 64 8.90 27.27 8.33 25.00 72.73 5–10 km 78 20.42 31.82 29.17 43.75 25.00 11–15 km 51 18.32 9.09 29.17 18.75 2.27 16–20 km 39 15.71 4.55 20.83 9.38 – 21–25 km 22 10.99 – – 3.13 – 26–30 km 13 5.76 4.55 4.17 – – 31–35 km 7 2.62 4.55 4.17 – – 36–40 km 4 1.05 4.55 4.17 – – More than 40 km 34 16.23 13.64 – – –
(n = 44), cyclists (n = 32), bus users (n = 24), and car passengers (n = 22;Table 1).
We invited all of the survey respondents to indicate their willingness to take part in completing an online travel diary. Of the 210 re-spondents who expressed their interest, we accepted the first 10 par-ticipants for each commuting mode (i.e., drivers, car passengers, bus users, cyclists, and pedestrians). Three of these participants did not fully complete their travel diary and the final sample consisted of 47 parti-cipants (41 female). Specifically, there were 10 drivers, 8 car passen-gers, 9 bus users, 10 cyclists, and 10 pedestrians in the final sample of commuters completing a travel diary.
2.2. Materials and procedure
The online survey included questions across 11 topic areas or sec-tions, arranged in a hierarchical structure. Each respondent received a minimum of 10 questions and, depending on their answers, a maximum of 198 questions. The question formats were a mix of check all that apply and forced-choice (see full question set inAppendix A). Of the 198 questions, two were filter questions (commuter mode type) de-termining the presentation of subsequent questions, ten were questions on demographic characteristics, four were questions on travel char-acteristics, and 183 were questions corresponding to the various mo-tivating factors (for additional details, seeAppendix A).
For the first section, we asked respondents to rate the difficulty of using each of the commuting modes available to them using a 5-point scale (1 = not difficult to 5 = very difficult) (adapted fromBerman and Cheng, 2001). Next, we asked respondents two questions on the per-ceived status associated with the commuting modes on a 7-point scale (1 = not at all to 7 = very much) (adapted from (Nelissen and Meijers, 2011). In the third section, we asked eight items (adapted from
Bamberg et al., 2003a) in which respondents rated their attitudes, subjective norms, PBC, and intentions to use each of the commuting modes available to them. For these items, a 5-point semantic differ-ential response scale was used.
For the next two sections (4 and 5), we asked questions pertaining only to respondents’ answers to the filter question “Which of these modes have you used for daily commute to and from work or school?” InSection 4respondents rated their satisfaction with using each com-muting mode using the 9-item Satisfaction with Travel Scale (STS; (Ettema et al., 2011), on a 9-point semantic differential scale. The Cronbach’s alpha reported byEttema et al. (2011)ranged from 0.84 to 0.91 across the three subscales of the STS. In Section 5, respondents rated the perceived autonomy and prestige associated with each mode across five 5-point Likert scale items (1 = strongly disagree to 5 = strongly agree) (adapted fromEllaway et al., 2003).
As an indication of the role of habit in their choice of commuting mode, Section 6 asked the respondents their frequency of using the travel modes for 10 destinations (e.g., going to the supermarket, visiting a friend in a nearby town), (adapted from Verplanken et al., 1994) Response-Frequency (RF) measure. Section 7 measured pro-ecological beliefs using the 15 item New Ecological Paradigm (NEP) scale (Dunlap et al., 2000). The NEP scale asks respondents to rate their agreement with a series of statements (on a 5-point Likert scale), seven of which are reverse scored to arrive at an overall score indicative of pro en-vironmental beliefs. The NEP scale has a Cronbach’s alpha of 0.80 (Dunlap et al., 2000). Section 8 contained nine 5-point Likert scale items from the Maximising Tendency Scale (MTS;Diab et al., 2008) to assess the extent to which the respondents try to optimise their deci-sions, or are willing to satisfice, making decisions that are just good enough.Diab et al. (2008)reported a Cronbach’s alpha of 0.80 for this scale.
Section 9 presented the Risk-Taking Index (RTI; Nicholson et al., 2005), designed to measure respondents’ tendencies to engage in or avoid risky behaviour (including driving fast and cycling without a helmet) across 28 items (5-point scale, 1 = never to 5 = very often). The
Cronbach’s alpha reported byNicholson et al. (2005)for all domain subscales ranged from 0.80 to 0.88. In Section 10, respondents provided details of their commute such as estimates of commuting time and distance and in Section 11, they provided details of their demographic and socio-economic characteristics (e.g., gender, age, occupation, etc., seeAppendix A).
In the travel diary, we asked respondents for details of each leg of all of their trips taken in a week. They completed one diary entry per day for 7 consecutive days (Monday to Sunday). We first asked how many trip legs (i.e., journeys between stops) they made on that day. The diary format allowed for a possible 15 trip legs each day. If a respondent made more than 15 trip legs, we asked them two more questions on the exact number of trip legs they made in that day and the mode they used most often for those trip legs. For each trip leg, we asked respondents 16 questions on their commute details such as their starting location, destination, estimated distance, estimated time, purpose, and the travel mode used for that trip leg. Out of the 16 questions, three were open-ended and 13 were multiple choice questions.
2.3. Analysis
To assess the role of habit we used the response frequency items and summed the number of times out of 10 that respondents chose their typical commuting mode for the destinations listed in the RF items. The possible range of scores was 0 (weak) to 10 (strong) habitual use. These scores were normally distributed with majority of the scores lying close to the mean (M = 5.45, 95% CI [5.13, 5.76], range = 9.00).
As a measure of respondents’ economic decision-making tendency, we averaged their scores on the RTI and the MTS. Our preliminary analysis indicated a negative correlation between scores on the RTI and the MTS. Therefore, we reverse-scored the items on the MTS before averaging the scores. The possible range of scores was 1 (low) to 5 (high) economic making tendency. The scores of economic decision-making tendency were normally distributed (M = 1.91, 95% CI [1.88, 1.94], range = 3.12).
To measure respondents’ social norms towards their travel modes, we averaged their scores on the two social-norm items fromSection 3
relevant to their mode choice. The two items were “Most people who are important to me would support the use of [travel mode] for daily commute next time” and “Most people who are important to me think that I should use [travel mode] for daily commute next time”. The possible range of scores was from 1 (weak) to 5 (strong) social norms. The data were negatively skewed, such that most of the commuters had positive social norms associated with their typical mode (M = 4.21, 95% CI [4.10, 4.31], range = 4.00). We obtained separate Cronbach’s alpha values for each commuter: drivers = 0.78, car passengers = 0.84, bus users = 0.97, cyclists = 0.87, and pedestrians = 0.90).
As a measure of respondents’ perceived status from using their travel modes, we averaged their scores on the two perceived status (Section 2) and the three perceived prestige (Section 5) items relevant to their mode choice. The possible range of scores was from 1 (low) to 5 (high) perceived status. Commuters’ scores of perceived status were normally distributed (M = 2.55, 95% CI [2.45, 2.64], range = 4.14).
To measure respondents’ derived pleasure from using their travel modes, we averaged their scores on the two attitude (Section 3), two perceived autonomy (Section 5), and nine STS (Section 4) items re-levant to their mode choice. We converted their average STS scores to a 5-point Likert scale to maintain consistency with the attitude and per-ceived autonomy scales. Respondents’ possible range of scores was from 1 (not pleasurable) to 5 (very pleasurable). The ratings of pleasure were normally distributed (M = 3.70, 95% CI [3.63, 3.77], range = 3.12). We also obtained separate Cronbach’s alpha values for each commuter: drivers = 0.22, car passengers = 0.59, bus users = 0.80, cy-clists = 0.31, and pedestrians = 0.50.
We averaged respondents’ scores on the two items for PBC (Section 3) and the perceived difficulty (Section 1) item relevant to their mode
choice to obtain their perceived-ease-of-using their respective travel modes. The two PBC items were “For me to use [travel mode] for daily commute next time would be 1 (difficult) or 5 (easy)” and “My freedom to use [travel mode] for daily commute next time is 1 (low) or 5 (high)”. We reverse-coded respondents’ perceived difficulty scores because the items in the perceived difficulty measure ranged from low perceived difficulty (i.e., easy) to high perceived difficulty (i.e., not easy), whereas the items in the PBC ranged from low PBC (i.e., not easy) to high PBC (i.e., easy). The possible range of scores for the perceived-ease-of-use was from 1 (not easy to use) to 5 (very easy to use). These scores were negatively skewed, such that most of the commuters per-ceived it was easy for them to commute using their usual mode (M = 4.72, 95% CI [4.66, 4.78], range = 2.50). We obtained separate Cronbach’s alpha values for each commuter: drivers = 0.63, car pas-sengers = 0.76, bus users = 0.66, cyclists = 0.53, and pedes-trians = 0.52.
To measure respondents’ ecological beliefs, we averaged their scores on the 15 items of the NEP (Section 7) after reverse-scoring the seven even-numbered items so that high scores on these items corresponded to pro-ecological beliefs. The possible range of scores was from 1
(negative) to 5 (positive) ecological beliefs. The data were normally distributed (M = 3.77, 95% CI [3.70, 3.83], range = 2.73.
We also measured respondents’ intentions to use their respective travel modes as an outcome variable in the regression analyses. We measured their intentions by averaging their scores on the two inten-tion items fromSection 3: “My intention to use [travel mode] for daily commute next time is 1 (weak) or 5 (strong)” and “I intend to use [travel mode] for daily commute next time” to which respondents answered 1 (unlikely) to 5 (likely)”. The possible range of scores for intention to use their travel mode was from 1 (weak) to 5 (strong). The intention scores were negatively skewed, such that most commuters had strong inten-tions to choose their usual mode for their next commute (M = 4.69, 95% CI [4.61, 4.77], range = 4.00. Where we identified missing values (21 responses) we replaced them with the respective mean scores of each commuter type for that particular measure to maintain the largest possible data set.
To examine respondents’ commitment towards their travel mode for their daily commute, we computed their percentages of total number of work or school trip legs and non-work or non-school trip legs involving their choice of travel mode over the week. In each daily entry, when a
Table 2
Parameter Values and Coefficients of Predictors from Multiple Linear Regression of Commuters’ Intentions to Use Their Travel Modes.
Predictor variables Coefficients Analysis of Variance
b SE 95% CI for b β p F-value df p R2 adj. R2
Drivers (n = 191) 5.87 7, 183 0.000 0.18 0.15 Constant (Intercept) 2.67 0.64 1.40, 3.93 – 0.000 Social norms 0.21 0.04 0.13, 0.30 0.36 0.000 Ease of use 0.19 0.09 0.01, 0.37 0.15 0.04 Habit 0.02 0.02 −0.01, 0.05 0.08 0.23 Ecological beliefs 0.08 0.07 −0.06, 0.22 0.08 0.26 Pleasure 0.01 0.07 −0.12, 0.15 0.01 0.85 Economic decision-making −0.07 0.14 −0.35, 0.20 −0.04 0.61 Perceived Status −0.02 0.05 −0.12, 0.08 −0.04 0.64 Passengers (n = 22) 4.12 7, 14 0.01 0.67 0.51 Constant (Intercept) −0.66 2.32 −5.63, 4.32 – 0.78 Social norms 0.67 0.29 0.04, 1.31 0.63 0.04 Pleasure 0.35 0.40 −0.50, 1.20 0.17 0.39 Ease of use 0.19 0.32 −0.49, 0.88 0.13 0.56 Ecological beliefs 0.60 0.46 −0.84, 1.15 0.10 0.74 Economic decision-making 0.11 0.59 −1.16, 1.39 0.04 0.85 Perceived Status −0.13 0.29 0.04, 1.31 −0.09 0.67 Habit −0.09 0.11 −0.32, 0.15 −0.21 0.43 Bus users (n = 24) 1.84 7, 16 0.15 0.45 0.20 Constant (Intercept) 1.53 2.41 −3.58, 6.63 – 0.54 Social norms 0.26 0.19 −0.13, 0.66 0.37 0.18 Pleasure 0.46 0.42 −0.43, 1.35 0.33 0.29 Economic decision-making 0.84 0.85 −0.96, 2.64 0.23 0.34 Habit 0.01 0.10 −0.21, 22 0.01 0.95 Ease of use −0.03 0.30 −0.65, 0.60 −0.02 0.93 Perceived Status −0.10 0.27 −0.68, 0.47 −0.08 0.71 Ecological beliefs −0.22 0.34 −0.93, 0.49 −0.14 0.52 Cyclists (n = 32) 3.27 7, 24 0.01 0.49 0.34 Constant (Intercept) 2.07 1.78 −1.61, 5.74 – 0.26 Ease of use 0.86 0.25 0.35, 1.37 0.64 0.002 Ecological beliefs 0.10 0.19 −0.28, 0.49 0.09 0.60 Habit 0.02 0.05 −0.08, 0.12 0.07 0.68 Social norms −0.03 0.14 −0.31, 0.26 −0.03 0.86 Perceived Status −0.05 0.13 −0.32, 0.22 −0.06 0.72 Pleasure −0.18 0.28 −0.75, 0.40 −0.10 0.53 Economic decision-making −0.36 0.34 −1.07, 0.35 −0.16 0.30 Pedestrians (n = 44) 11.33 7, 36 0.000 0.69 0.63 Constant (Intercept) −1.57 1.10 −3.79, 0.66 – 0.16 Ease of use 0.78 0.25 0.28, 1.29 0.46 0.004 Pleasure 0.61 0.19 0.21, 1.00 0.40 0.004 Habit 0.08 0.06 −0.04, 0.21 0.13 0.19 Perceived Status 0.07 0.11 −0.15, 0.30 0.07 0.50 Economic decision-making 0.02 0.31 −0.61, 0.66 0.01 0.94 Social norms 0.01 0.15 −0.30, 0.31 0.01 0.96 Ecological beliefs −0.10 0.14 −0.40, 0.19 −0.07 0.47
respondent chose either work or education as the purpose of their trip, we considered their response as a ‘work or school trip leg’ and con-sidered every other trip purpose as ‘non-work or non-school’ trip leg. For bus users, we excluded ‘change-to-other-mode’ trip legs (i.e., walking to and from bus stops) from their total number of trip legs in a week to avoid a low percentage of work or school trip legs involving bus use. High percentages indicate strong commitment towards their travel mode (i.e., large number of trip legs involving their travel mode), whereas low percentages indicate weak commitment. For work or school trip legs, we found a mean of 78.33 (95%CI [70.60, 86.05], range = 100) and for non-work or non-school trip legs, we found a mean of 49.88 (95%CI [42.63, 58.13], range = 100). The commitment data for work or school trip legs were negatively skewed (i.e., many commuters were highly committed to their usual mode) but the com-mitment data for non-work or non-school trip legs were normally dis-tributed.
3. Results
3.1. Motivating factors predicting intentions to use travel mode for daily commute
In order to identify the motivating factors predicting intentions to use particular travel modes we carried out a multiple regression ana-lysis for each type of commuter. Our regression analyses showed that the social norms measure was a significant predictor of drivers’ and passengers’ intentions to use a car for their commute. Perceived ease-of-use was a significant predictor of drivers’, cyclists’ and pedestrians’ intentions (to drive, cycle, and walk, respectively) and pleasure was also a significant predictor of pedestrians’ intentions to walk for their daily commute. No other motivating factors significantly predicted the commuters’ intentions to use their respective modes and none of mo-tivating factors predicted bus users’ intentions to use the bus for their daily commute (Table 2).
Table 3
Parameter Values and Coefficients of Each Factor from the one-way ANOVA.
Motivating Factors Coefficients Analysis of Variance
N Mean SE 95% CI F-value p η2 Social Norms 2.24* 0.07 0.03 Pedestrians 44 4.49 0.12 4.25, 4.73 Drivers 191 4.24 0.07 4.11, 4.37 Cyclists 32 4.09 0.17 3.76, 4.42 Passengers 22 3.94 0.20 3.54, 4.34 Bus users 24 3.85 0.27 3.33, 4.37 Ease of Use 4.55* 0.003 0.10 Drivers 191 4.82 0.03 4.76, 4.88 Pedestrians 44 4.69 0.07 4.55, 4.83 Cyclists 32 4.68 0.10 4.49, 4.87 Bus users 24 4.39 0.16 4.08, 4.70 Passengers 22 4.36 0.14 4.08, 4.64 Pleasure 27.10 0.000 0.35 Cyclists 32 4.49 0.07 4.35, 4.63 Pedestrians 44 4.00 0.08 3.83, 4.17 Passengers 22 3.92 0.10 3.72, 4.12 Bus users 24 3.51 0.13 3.25, 3.77 Drivers 191 3.49 0.04 3.41, 3.57 Habit 72.02* 0.000 0.62 Drivers 191 6.74 0.17 6.40, 7.08 Passengers 22 5.27 0.51 4.28, 6.26 Cyclists 32 4.03 0.42 3.20, 4.86 Bus users 24 2.83 0.39 2.07, 3.59 Pedestrians 44 2.40 0.20 2.01, 2.79 Economic decision-making 2.55 0.04 0.03 Drivers 191 1.95 0.02 1.91, 1.99 Cyclists 32 1.91 0.17 1.58, 2.24 Pedestrians 44 1.86 0.12 1.62, 2.10 Passengers 22 1.84 0.07 1.71, 1.97 Bus users 24 1.79 0.27 1.27, 2.31 Status 2.17 0.07 0.03 Pedestrians 44 2.85 0.13 2.59, 3.10 Cyclists 32 2.68 0.16 2.36, 3.00 Drivers 191 2.48 0.06 2.36, 2.60 Passengers 22 2.44 0.15 2.12, 2.75 Bus users 24 2.41 0.15 2.09, 2.73 Ecological Beliefs 1.36 0.25 0.02 Drivers 191 4.82 0.03 4.76, 4.88 Pedestrians 44 4.69 0.07 4.55, 4.83 Cyclists 32 4.68 0.10 4.49, 4.87 Passengers 22 4.36 0.14 4.08, 4.64 Bus users 24 4.39 0.16 4.08, 4.70
Note: *We used Welch’s F as the ANOVA statistic because the assumption of homogeneity of variance was violated. Where we used Welch’s F, we used the Games-Howell post-hoc analysis for pairwise comparisons.
3.2. Differences in motivating factors between commuters
Another way of looking at the relative influence of the seven mo-tivating factors is to compare them across the different commuter types. To do this, we conducted a one-way, between-subject, ANOVA for each factor, where the independent variable was commuter group with five levels (drivers, passengers, bus users, cyclists, and pedestrians). From
Table 3, we see that commuters were significantly different from each other in terms of the ease-of-use, pleasure, economic decision-making, and habit factors. They were not significantly different from each other in terms of the social norms, perceived status, and ecological beliefs factors.
All the commuters had medium to high scores for the social norms measure. For the ease-of-use factor, post-hoc pairwise comparisons (Games-Howell) showed that drivers perceived using their travel mode to be easier than car passengers perceived theirs. Post-hoc pairwise comparisons (Bonferroni-adjusted) for the ANOVA on derived pleasure from using travel modes showed that cyclists’ found using their travel mode (i.e., cycling) more pleasurable than all the other commuters did. Passengers found using their travel mode (i.e., car as passenger) more pleasurable than drivers did, whereas pedestrians found using their travel mode (i.e., walking) more pleasurable than bus users did.
As for habit, post-hoc pairwise comparisons (Games-Howell) de-monstrated that drivers have a stronger habit of using their travel mode
than bus users, cyclists, and pedestrians do. Passengers have a stronger habit of using their travel mode than bus users and pedestrians do, whereas cyclists have stronger habit of cycling than pedestrians do towards walking. Although the ANOVA for economic decision-making indicated an overall significant difference, there were no pairwise dif-ferences. All commuters had low scores for the perceived status factor and high scores on the ecological beliefs factor.
3.3. Mode commitment
Another interesting perspective on commuters’ use of a particular mode was revealed by examining the degree to which they used one mode, and only that mode for their travel; in other words their com-mitment to a particular travel mode. So, using the 1-week travel diary data, we identified the commuters’ percentage of trip legs involving their travel mode and alternative modes. We considered commuters who used their travel mode most frequently in a week to have higher commitment towards their travel mode. Interestingly, we found that all commuters were committed to their travel modes because they used their respective modes most frequently compared to alternative modes for daily commute.
Fig. 1shows the five commuter groups’ mean percentages of work or school trip legs involving their respective mode choice. Drivers drove most of the time in the week (92.47%) and walked for the remaining
Fig. 1. Mean percentages of each commuter groups’ work or school trip legs using various modes over a week.
Fig. 2. Mean percentages of respondents’ work or school trip legs and non-work or non-school trip legs using their choice of travel mode over a week for five
trip legs (7.53%). Passengers used the car as a passenger most of the time (60.94%), followed by walking (29.21%) and taking the bus (4.29%). Most of pedestrians’ trip legs involved walking (88.00%) and the rest involved driving (8.67%) and using the car as passenger (3.33%). Bus users used the bus frequently (68.89%) and cyclists cycled for most of their trip legs (69.08%). Car passengers, bus users, and cyclists used all other modes occasionally, whereas drivers and pedes-trians used their travel modes for their most of their daily commute trips.
We then wondered whether commuters’ commitment levels to their mode choice would be different for daily commute trips (i.e., non-work or non-school trips). Therefore, we computed each commuter groups’ mean percentages of non-work or school trip legs involving their travel mode in a week.Fig. 2shows that all commuters, except car passengers, used their travel mode more often for work or school trip legs compared to non-work or non-school trip legs. However, the re-peated measures t-tests (Table 4) found that the differences were only significant (Bonferroni-adjusted) for bus users, cyclists, and pedestrians, such that these commuters were more likely to use their travel modes for daily commute trips rather than non-daily commute trips. Drivers and car passengers were equally likely to use the car for both types of trips.
4. Discussion
Strongly committed commuters are less likely to use other travel modes for any of their trips (Simma and Axhausen, 2001). In our study, drivers were the most committed towards their travel mode, such that they drove their car for the greatest proportion of their daily trips. Few drivers in our study used other modes, with the exception of a small percentage of trip legs that involved walking. Buehler and Hamre (2016)also reported that walking was the only non-car mode used by the majority of car users. Other researchers have described commitment as either mono-modal when only one travel mode is used for most trips or multi-modal when multiple travel modes are used (Heinen and Chaterjee, 2015; Kuhnimhof et al., 2006; Nobis, 2007). In these terms,
Kuhnimhof et al. (20060 reported that car drivers can usually be characterised as mono-modal.
Like drivers, the pedestrians in our study also showed high com-mitment and used fewer travel modes compared to car passengers, bus users, and cyclists who used a variety of modes for their trips. Certainly part of this pattern may be due to situational constraints, such as when bad weather causes cyclists to use weather-sheltered modes (i.e., car or bus), or very short trip distances that occasion walking. Similarly, non-routine trips often involve a range of destinations (e.g., church, su-permarket, gym) encouraging the use of a variety of travel modes. In other words, non-routine or unfamiliar trips may occasion alternative mode choices to be considered (Carrel et al., 2011; Wood et al., 2005). Looking beyond mode commitment, our results indicated that there are different motivating factors underlying different types of commu-ters’ mode choices. Social norms was the best predictor of drivers’ and passengers’ intentions. It is also noteworthy that there were no sig-nificant differences in the social norm scores of any of the commuter
types, reporting relatively neutral to positive social norms for their respective travel modes. Previous studies have found that family members and friends can be a source of inspiration to drivers through observations of car use (Jopson, 2004; Noblet et al., 2014) or con-versations around car use (Haustein et al., 2009). This was also the case for car passengers (Teal, 1987) and pedestrians (Schneider, 2011). Therefore, interventions aimed to influence the attitudes of commuters towards sustainable travel modes, could also influence the attitudes of the people in commuters’ social networks (e.g., family members, friends, co-workers, and neighbours).
Ease-of-use was a significant predictor of drivers’, cyclists’, and pedestrians’ intentions to use their travel modes. In contrast to social norms, there were significant differences between the commuter groups in terms of their perceptions of ease of using their modes, with drivers reporting the highest score for this factor. Drivers may find it con-venient to use the car because they can carry out their daily responsi-bilities (e.g., transport children from day care, and shopping) (Maxwell and Miller, 2001), and park their cars easily at places that offer con-venient parking such as workplaces, universities, and shopping malls (Mackett, 2003). The cyclists in our study also had high scores for this factor. Cyclists in previous studies have demonstrated positive self-ef-ficacy towards cycling (De Gues et al., 2008; Venkatesh and Davis, 1996) and some claimed that having the necessary facilities that com-plemented their cycling behaviour, such as bicycle lockers and shower rooms made it easier for them to cycle to work or school (Dickinson et al., 2003; Hunt and Abraham, 2007). The pedestrians in our study also reported high perceived ease-of-use, possibly a consequence of the shorter distances of their commutes, an average of less than 5 km a day (Table 1).
Pedestrians’ pleasure from walking was also a significant predictor of their intentions to walk for their daily commute. Indeed this intrinsic pleasure in the activity has been previously shown to be important to many pedestrians (Gatersleben and Uzzell, 2007). As with the ease-of-use ratings, we found significantly different pleasurability ratings for the different commuter groups. Cyclists were more likely to find using their travel mode pleasurable than all other commuters. Cycling not only provides physical and mental health benefits, but it also provides fun, such that cyclists can enjoy the view, breathe fresh air, and feel accomplished when they have gone on a long ride (Schneider, 2011). Drivers had the lowest score on this measure, albeit their mean score indicated neutral feelings from driving. One option for encouraging drivers to consider other mode choices may thus be to emphasise the less pleasurable aspects of car use (e.g., traffic congestion and stress) and make the pleasurable aspects of active commuting more apparent (Hennessy and Wiesenthal, 1999; Koslowsky et al., 1995).
Although previous researchers have observed economic influences on mode choice decisions (Beirao and Carbral, 2007; Cervero, 2002; Corpuz, 2007) we were unable to detect any significant influences in our study and there were no appreciable differences between commuter types on these measures. We also examined status as a predictor of commuters’ intentions to use their travel modes. Based on previous research that identified status as an important aspect in car use (e.g.,
Steg et al., 2001; Steg, 2005), we expected to see an influence of
Table 4
Parameter Values and Coefficients for Repeated Measures t tests Comparing Percent of Trip Legs for Daily vs. Non-daily Commute. Percent of work or school trip legs Percent of non-work or school trip legs
M 95% CI M 95% CI t p d Drivers 92.47 85.30, 99.64 72.90 52.86, 92.92 1.58 0.149 0.79 Car passengers 60.94 31.63, 90.25 62.10 23.19, 64.27 −0.12 0.905 0.04 Bus users 68.89 50.17, 87.61 34.15 26.29, 37.11 3.78 0.005* 1.76 Cyclists 69.08 52.56, 85.60 32.83 22.58, 43.08 5.75 0.000* 1.63 Pedestrians 88.00 76.98, 99.02 45.50 29.8, 61.20 5.24 0.001* 1.94
perceived status on commuters’ intentions to use their travel modes, particularly in drivers and car passengers. All the commuters in our study had comparably low scores on the perceived-status measure, and this factor did not predict any of their intentions to use their respective travel modes.
Although there is a general consensus that pro-environmental atti-tudes or beliefs are associated with sustainable travel behaviour (Atasoy et al., 2012; Kahn and Morris, 2009), our measure of ecological beliefs was not a significant predictor of any of the five types of commuters’ intentions to use their respective travel modes and there were no sig-nificant ecological belief differences between the commuters. Similarly, other researchers found that different types of commuters had similar environmental attitudes (Thomas and Walker, 2015) and their pro-en-vironmental beliefs neither increased their bus use (Heath and Gifford, 2002) nor reduced their car use (Fujii, 2006). We believe that social desirability bias may have influenced our findings, such that re-spondents answered the ecological belief-related questions so as to appear to have pro-ecological beliefs. Some individuals may overstate their frequency of performing pro-environmental behaviour (e.g., using green travel modes and recycling) to appear pro-environmental (Thogersen and Olander, 2006) or report that they are pro-environ-mental but actually do nothing to reflect their attitude (Sargisson and McLean, 2015). Therefore, interventions aimed at enhancing commu-ters’ ecological beliefs around active commuting may not be very ef-fective, particularly in our group of commuters because they already have positive ecological beliefs. Further research could potentially look into how to manifest commuters’ ecological beliefs into actions that actually support their beliefs and include a social desirability scale to test for this bias explicitly.
The final factor we examined was the amount of previous experi-ence commuters had using their preferred mode (habit strength or re-sponse frequency). Although there were significant differences between the commuter groups, the response frequency measure was not a sig-nificant predictor of any of the commuters’ intentions to use their re-spective travel modes. Although it may seem counterintuitive to use a response frequency measure of what is often an automatic behaviour (i.e., habit) to predict a controlled, conscious measure of choice (i.e., intention), consistent with previous research drivers had the highest prior experience of using their travel mode and pedestrians the lowest (Bamberg et al., 2003a,b; Ouellette and Wood, 1998).
Despite our interesting findings, our study had several limitations. We used commuters’ intentions to use their respective travel modes as the proxy outcome variable in the regression analyses of the seven motivating factors, instead of their actual behaviour. Commuters’ in-tentions to use their travel mode may not truly reflect the extent to which they actually use the travel modes, as there is evidence of the intention-behaviour gap, such that medium-to-large-sized change in intentions lead to only small-to-medium-sized changes in actual beha-viour (Sheeran, 2002; Webb and Sheeran, 2006). So, we asked the re-spondents to complete a self-report travel diary under the assumption that their travel diary responses would more closely resemble their actual commuting behaviour. Many travel behaviour researchers have used intentions as an outcome measure (e.g.,Bamberg, 2000; Bamberg et al., 2003a;Matthies et al., 2002) and have used travel diaries to examine commuters’ actual behaviour (e.g., Bamberg et al., 2003b; Garvill et al., 2003; Schlich and Axhausen, 2003).
A second sort of issue arose in the context of our mode commitment results. Specifically, the travel diary participants were comprised of the first ten volunteers from each commuter group, a self-selected sample
that may have had high levels of commitment to their commuting practices with some influence on their willingness to complete the travel diary, thus resulting in high commitment levels. This also meant that some of the travel diary participants were from outside the regional sample (11 out of the 47). The potential for self-selection bias, as well as the gender imbalance in our sample (approximately 75% female) was present to some degree in both the online survey and the travel diary portions of the study and represents an ongoing methodological chal-lenge for this research going forward.
Another issue to be addressed going forward would be to use techniques such as structural equation modelling to contrast the com-peting influences for observed mode choice decisions, instead of stated intentions to choose a particular commuting mode. This analysis ap-proach was not possible in the present study due to the wording of our mode choice questions and the limited number of respondents in some of our commuter groups (Kline, 2011, suggested a sample size of 100 or more for each observation category). In this context, some additional differentiation of types of drivers could also be useful for analysis. In the present study, the sample of drivers was very heterogeneous as reflected in both the demographic variables and in the low predictive values observed in the regression analyses of their mode choice inten-tions (relative to some of the other commuter groups). This differ-entiation could lead to a better understanding of drivers’ commuting choices and identification of the best ways to nudge them to make more sustainable choices.
Finally, our habit strength measure may not adequately reflect commuters’ habit strength of using their respective travel modes be-cause we did not impose a time pressure on our respondents for the RF measure which, according toVerplanken et al. (1994), was essential to capture the element of automaticity of commuters’ habitual use of their travel mode.
Overall, our study has created the potential for further studies to understand the influence of the motivating factors on commuters’ mode choice in depth and to incorporate them in interventions aimed at re-ducing the use of cars and increasing the use of sustainable travel modes. The commuters in our study reported conflicting motivating factors related to their intentions to use their respective travel modes, which provides the opportunity to ‘nudge’ them in different ways to use more sustainable travel modes without forbidding any choice or sig-nificantly altering the economic rewards of any choice (see Nudge Theory;Thaler and Sunstein, 2008). We are optimistic that these tai-lored and well-tested nudge techniques can reduce car use and even-tually reduce the emission of poisonous gases that pose a threat to the environment and its inhabitants.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influ-ence the work reported in this paper.
Acknowledgements
This research was made possible by funding from the School of Psychology at the University of Waikato. We would like to thank our participants for taking part in our study and the organisations that as-sisted us in recruiting the participants for our study. We would also like to thank Jes Leov for her assistance in developing the questionnaire and travel diary.
Appendix A
Questions Used in the Online Survey
Survey Questions Response scale Note
Filter question #1:
During your daily commute to work or school, which of the following modes are available to you? (For your understanding, a mode is available to you when you have unrestricted access to it regardless whether you use the mode or not.
1. Private car as driver 2. Private car as passenger 3. Public transport: Bus 4. Bicycle
5. Walking 6. I do not commute
Respondents who selected ‘I do not com-mute’ advanced to the demographic ques-tions (Section 10)
Section 1
Measures of perceived difficulty
1. Please rate how difficult would it be for you to use [travel mode] for daily commute to and from work or school.
5-point scale
1 = not difficult to 5 = very difficult All commuters respond to one item for eachmode that they selected in Filter question #1
Section 2
Measures of perceived status
1. If I used [travel mode] for daily commute to and from work or school I will have high status
2. If I used [travel mode] for daily commute to and from work or school I will be respected
7-point scale
1 = not at all to 7 = very much All commuters respond to two items for eachmode that they selected in Filter question #1
Section 3
Measures of attitude
For me, to use [travel mode] for daily commute next time would overall be: 1. Bad or good
2. Unpleasant or pleasant
Measures of social norms
1. Most people who are important to me would support the use of [travel mode] for daily commute next time.
2. Most people who are important to me think that I should use [travel mode] for daily commute next time.
Measures of perceived behavioural control
1. For me to use [travel mode] for daily commute next time would be: difficult or easy
2. My freedom to use [travel mode] for daily commute next time is: low or high
Measures of intentions
1. My intention to use [travel mode] for daily commute next time is: weak or strong
2. I intend to use [travel mode] for daily commute next time: unlikely or likely
5-point semantic differential scale5-point scale 1 = disagree to 5 = agree5-point semantic differen-tial scale5-point semantic differendifferen-tial scale
All commuters respond to eight items for each mode that they selected in Filter ques-tion #1
Filter question #2:
Which of these modes have you used for daily commute to and from work or school?
1. Private car as driver 2. Private car as passenger 3. Public transport: Bus 4. Bicycle
5. Walking 6. None of the above
Respondents who selected ‘None of the above’ advanced to the demographic ques-tions (Section 10)
Section 4
Measures of satisfaction with travel
1. Travelling on [travel mode] is: very hurried or very relaxed
2. When I travel in [travel mode], I’m: very worried I would not be in time or very confident I would be in time
3. When I travel in [travel mode], I’m: very stressed or very calm 4. When I travel in [travel mode], I’m: very tired or very alert 5. When I travel in [travel mode], I’m: very bored or very enthusiastic 6. When I travel in [travel mode], I’m: very fed-up or very engaged 7. Travelling on [travel mode] was the: worst or best I can think of 8. Travelling on [travel mode] had: very low standard or very high standard 9. Travelling on [travel mode] worked: very poorly or very well
9-point semantic differential scale All commuters respond to nine items for each mode that they selected in Filter ques-tion #2
Section 5
Measures of perceived autonomy
1. I can travel where I want, when I want by [travel mode] 2. I feel in control when I travel by [travel mode]
3. Travelling in [travel mode] fits in well with the routine of my daily life
Measures of perceived prestige
1. Most people would like to travel by [travel mode] like the way I do 2. When I travel by [travel mode] it makes me feel I’m doing well in life 5-point scale
1 = strongly disagree to 5 = strongly agree5-point scale 1 = strongly disagree to 5 = strongly agree
Perceived autonomy: All commuters respond to three items for each mode that they selected in Filter question #2
Perceived prestige: All commuters respond to two items for each mode that they selected in Filter question #2
Section 6
Measures of habit
Please indicate which travel mode you would choose for the following 10 destinations or trip purposes:
1. Summer excursion with friends to a lake 2. Visit a friend
3. Visit your parents who live 3 km away 4. Engage in sports
5. Stroll through the city 6. Evening visit to a bar 7. A trip on a nice day 8. Routine grocery shopping 9. Eat in a restaurant 10. 10. Go to the movies
Respondents select one of the five modes (i.e., car as driver, car as passenger, bus, bicycle, or walk)
All commuters respond to all ten items
Section 7
Measures of ecological beliefs
1. We are approaching the limit of the number of people the earth can support 2. Humans have the right to modify the natural environment to suit their needs
(R)
3. When humans interfere with nature it often produces disastrous conse-quences
4. Human ingenuity will ensure that we do NOT make the earth unliveable (R) 5. Humans are severely abusing the environment
6. The earth has plenty of natural resources if we just learn how to develop them (R)
7. Plants and animals have as much right as humans to exist
8. The balance of nature is strong enough to cope with the impacts of modern industrial nations (R)
9. Despite our special abilities humans are still subject to the laws of nature 10. The so-called “ecological crisis” facing humankind has been greatly
ex-aggerated (R)
11. The earth is like a spaceship with very limited room and resources 12. Humans were meant to rule over the rest of nature (R)
13. The balance of nature is very delicate and easily upset
14. Humans will eventually learn enough about how nature works to be able to control it (R)
15. If things continue on their present course, we will soon experience a major ecological catastrophe
5-point scale 1 = strongly disagree to 5 = strongly
agree All commuters respond to all 15 items
Section 8
Measures of maximising tendency
1. No matter what it takes, I always try to choose the best thing 2. I don’t like having to settle for “good enough”
3. I am a maximiser
4. No matter what I do, I have the highest standards for myself 5. I will wait for the best option, no matter how long it takes 6. I never settle for second best
7. I am uncomfortable making decisions before I know all of my options 8. Whenever I’m faced with a choice, I try to imagine what all the other
possibilities are, even ones that aren’t present at the moment 9. I never settle
5-point scale 1 = strongly disagree to 5 = strongly
agree All commuters respond to all nine items
Section 9
Measures of risk-taking tendency
We are interested in everyday risk-taking. Please could you tell us if any of the following have ever applied to you, now and in your adult past?
1. Rock Climbing 2. Scuba Diving 3. Smoking 4. Poor diet
5. High alcohol consumption
6. Quitting a job without another to go to 7. Gambling
8. Risky investments 9. Fast driving
10. City cycling without a helmet 11. Standing for election
12. Publicly challenging a rule or a decision
5-point scale
1 = strongly disagree to 5 = strongly agree All commuters respond to all 12 items twice(for present and past)
Section 10
Travel characteristics
1. Most frequently used mode 2. Mode use period
3. Estimated commuting time in a day 4. Estimated commuting distance in a day
Section 11 Demographic characteristics 1. Gender 2. Age 3. Occupation 4. Education level 5. Household type 6. Annual household income 7. Place of residence 8. Moving status 9. Period since last move
Nine questions
Appendix B. Supplementary data
Supplementary data to this article can be found online athttps://doi.org/10.1016/j.tbs.2019.10.007.
References
Aarts, H., Dijksterhuis, A., 2000. The automatic activation of goal-directed behaviour: the case of travel habit. J. Environ. Psychol. 20, 75–82.
Ajzen, I., 1991. The theory of planned behaviour. Org. Behav. Human Decis. Processes 20, 179–211.
Anable, J., Gatersleben, B., 2005. All work and no play? The role of instrumental and affective factors in work and leisure journeys by different travel modes. TRA: Policy Pract. 39 (2–3), 163–181.
Atasoy, B., Glerum, A., Bierlaire, M., 2012. Attitudes Towards Mode Choice in Switzerland. Transport and Mobility Laboratory (TRANSP-OR), Lausanne, Switzerland Retrieved from https://transp-or.epfl.ch.
Bamberg, S., 2000. The promotion of new behaviour by forming an implementation in-tention: results of a field experiment in the domain of travel mode choice. J. Appl. Soc. Psychol. 30 (9), 1903–1922.
Bamberg, S., 2006. Is residential relocation a good opportunity to change people’s travel behaviour? Results from a theory-driven intervention study. Environ. Behav. 38 (6), 820–840.
Bamberg, S., Ajzen, I., Schmidt, P., 2003a. Choice of travel mode in the theory of planned behaviour: the roles of past behaviour, habit and reasoned action. Basic Appl. Soc. Psychol. 25 (3), 175–187.
Bamberg, S., Rolle, D., Weber, C., 2003b. Does habitual car use not lead to more re-sistance to change of travel mode? Transp. Plann. Technol. 30 (1), 97–108.
Beirao, G., Cabral, J.A.S., 2007. Understanding attitudes towards public transport and private car: a qualitative study. Transp. Policy 14, 478–489.
Beirao, G., Carbral, J.A.S., 2007. Understanding attitudes towards public transport and private car: a qualitative study. Transp. Policy 14, 478–489.
Berman, R., Cheng, L., 2001. English academic language skills: perceived difficulties by undergraduate and graduate students, and their academic achievement. Can. J. Appl. Linguist. 4 (1), 25–40.
Buehler, R., Hamre, A., 2016. An examination of recent trends in multimodal travel be-havior among American motorists. Int. J. Sustain. Transp. 10 (4), 354–364.
Cairns, S., Davies, A., Newson, C., Swiderska, C., 2002. Making travel plans work: Research report. Department of Transport, London, UK.
Cairns, S., Sloman, L., Newson, C., Anable, J., Kirkbride, A., Goodwin, P., 2008. Smarter choices: assessing the potential to achieve traffic reduction using 'soft measures'. Transp. Rev. 28 (5), 593–618.
Carrel, A., Vij, A., Walker, J.L., January 2011. Understanding multimodal behavior: in-dividual modality styles and their influence on travel behavior. In: Paper presented at the Transportation research Board 90th annual Meeting, Washington, WA.
Cervero, R., Sarmiento, O.L., Jacoby, E., Gomez, L.F., Neiman, A., 2009. Influences of built environments on walking and cycling: lessons from Bogotá. Int. J. Sustain. Transp. 3 (4), 203–226.
Cervero, R., 2002. Built environments and mode choice: toward a normative framework. TRD: Transp. Environ. 7 (4), 265–284.
Corpuz, G., 2007. Public transport or private vehicles: Factors that impact on mode choice. Paper presented at the Australasian Transport Research Forum (ATRF), Melbourne, Australia.
Davies, D.G., Halliday, M.E., Mayes, M., Pocock, R.L., 1997. Attitudes to Cycling: A Qualitative Study and Conceptual Framework. Transport Research Laboratory, Crowthorne.
De Gues, B., De Bourdeaudhuij, I., Jannes, C., Meeusen, R., 2008. Psychosocial and en-vironmental factors associated with cycling for transport among a working popula-tion. Health Educ. Res. 23 (4), 697–708.
Diab, D.L., Gillespie, M.A., Highhouse, S., 2008. Are maximisers really unhappy? The measurement of maximising tendency. Judgement Decis. Making 3 (5), 364–370.
Dickinson, J.E., Kingham, S., Copey, S., Houghie, D.J.P., 2003. Employer travel plans, cycling and gender: will travel plan measures improve the outlook for cycling to work in the UK? TRD: Transp. Environ. 8 (1), 53–67.
Dunlap, R.E., Liere, K.D.V., Mertig, A.G., Jones, R.E., 2000. Measuring endorsement of the New Ecological Paradigm: a revised NEP scale. J. Social Issues 56 (3), 425–442.
Eliasson, J., Hultkrantz, L., Nerhagen, L., Rosqvist, L.S., 2009. The Stockholm congestion -charging trial 2006: overview of effects. TRA: Policy Pract. 43, 240–250.
Ellaway, A., Macintyre, S., Hiscock, R., Kearns, A., 2003. In the driving seat: psychosocial benefits from private motor vehicle transport compared to public transport. TRF:
Traffic Psychol. Behav. 6, 217–231.
Ettema, D., Garling, T., Eriksson, L., Friman, M., Olsson, L.E., Fujii, S., 2011. Satisfaction with travel and subjective well-being: development and test of a measurement tool. TRF: Traffic Psychol. Behav. 14, 167–175.
Ettema, D., Friman, M., Garling, T., Olsson, L.E., Fujii, S., 2012. How in-vehicle activities affect work commuters' satisfaction with public transport. J. Transp. Geogr. 24, 215–222.
Flamm, B., 2009. The impacts of environmental knowledge and attitudes on vehicle ownership and use. TRD: Transp. Environ. 14 (4), 272–279.
Friman, M., Edvarsson, B., Garling, T., 2001. Frequency of negative critical incidents and satisfaction with public transport services (I). J. Retailing Consumer Serv. 8, 95–104.
Fujii, S., 2006. Environmental concern, attitude toward frugality, and ease of behavior as determinants of pro-environmental behavior intentions. J. Environ. Psychol. 26 (4), 262–268.
Fujii, S., Garling, T., 2003. Development of script-based travel mode choice after forced change. TRF: Traffic Psychol. Behav. 6, 117–124.
Gardner, B., 2009. Modelling motivation and habit in stable travel mode contexts. TRF: Traffic Psychol. Behav. 12, 68–76.
Garling, T., Schuitema, G., 2007. Travel demand management targeting reduced private car use: effectiveness, public acceptability and political feasibility. J. Social Issues 63 (1), 139–153.
Garvill, J., Marell, A., Nordlund, A., 2003. Effects of increased awareness on choice of travel mode. Transportation 30, 63–79.
Gatersleben, B., Uzzell, D., 2007. Affective appraisals of the daily commute: comparing perceptions of drivers, cyclists, walkers, and users of public transport. Environ. Behav. 39 (3), 416–431.
Haque, M.M., Chin, H.C., Debnath, A.K., 2013. Sustainable, safe, smart—three key ele-ments of Singapore’s evolving transport policies. Transp. Policy 27, 20–31.
Haustein, S., Klockner, C.A., Blobaum, A., 2009. Car use of young adults: the role of travel socialization. TRF: Traffic Psychol. Behav. 12, 168–178.
Heath, Y., Gifford, R., 2002. Extending the theory of planned behavior: predicting the use of public transport. J. Appl. Social Psychol. Market. 32 (10), 2154–2189.
Heinen, E., Chaterjee, K., 2015. The same mode again? An exploration of mode choice variability in Great Britain using the National Travel Survey. TRA: Policy Pract. 78, 266–282.
Hennessy, D., Wiesenthal, D., 1999. Traffic congestion, driver stress, and driver aggres-sion. Aggressive Behav. 25 (6), 409–423.
Holmberg, K., Anderson, P., Erdemir, A., 2012. Global energy consumption due to friction in passenger cars. Tribol. Int. 47, 221–234.
Hunecke, M., Blobaum, A., Matthies, E., Hoger, R., 2001. Responsibility and environment: ecological norm orientation and external factors in the domain of travel mode choice behavior. Environ. Behav. 33 (6), 830–852.
Hunt, J.D., Abraham, J.E., 2007. Influences on bicycle use. Transportation 34 (4), 453–470.
International Organization of Motor Vehicle Manufacturers (OICA), 2015. Vehicles In Use. Retrieved from.http://www.oica.net/category/vehicles-in-use/.
Jones, P.M., Sloman, L., 2003. Encouraging behavioural change through marketing and management: What can be achieved? Paper presented at the 10th International Conference on Travel Behaviour Research, Lucerne, Switzerland. Retrieved from http://citeseerx.ist.psu.edu/.
Jopson, A., 2004. The role of norms in mode choice. Paper presented at the 3rd International Conference on Traffic and Transportation Psychology, Nottingham, United Kingdom. Retrieved from http://eprints.whiterose.ac.uk/2501.
Kahn, M.E., Morris, E.A., 2009. Walking the walk: the association between community environmentalism and green travel behaviour. J. Am. Plann. Assoc. 75 (4), 389–405.
Kahneman, D., Tversky, A., 1979. Prospect theory: an analysis of decision under risk. Econometrica 47 (2), 263–291.
Kline, R.B., 2011. Principles and Practice of Structural Equation Modeling. The Guilford Press, New York, NY.
Koslowsky, M., Kluger, A., Reich, M., 1995. Commuting Stress: Causes, Effects and Methods of Coping. Plenum, New York, NY.
Kuhnimhof, T., Chlond, B., Von der Ruhren, S., 2006. Users of modes and multimodal travel behaviour: steps toward understanding travellers' options and choices. Transp. Res. Rec.: J. Transp. Res. Board 1985 (1), 40–48.
Mackett, R.L., 2003. Why do people use their cars for short trips? Transportation 30, 329–349.
